Studies are proposed to improve understanding of the structure and organization of declarative memory and the nature of memory disorders. The work is organized as three topics and involves 7 projects. Notable among the proposed projects is the extraordinary opportunity to carry out neurohistological analysis of the brains of two of our most interesting study patients (EP and AB), who passed away during the current funding period after we had studied them for 13 and 24 years, respectively. An additional salient aspect of the proposed work is the continuing opportunity to investigate memory by studying memory-impaired patients with bilateral lesions of the hippocampus or with larger medial temporal lobe lesions. Our standing population of amnesic study patients is one of the very few such populations available anywhere, and their pathology is the best characterized in terms of quantitative neuroanatomy. Neuroimaging studies of learning and memory in healthy volunteers are also proposed. Aside from the neurohistology project (A), I also propose 4 studies of the nature of recognition memory and the constructs of recollection and familiarity, with special attention to the functions of hippocampus and perirhinal cortex. B1. I propose to determine the shape of the function within hippocampus and perirhinal cortex that relates fMRI activity to memory strength; B2. I propose to assess neural activity associated with Remembering and Knowing, but after matching for memory strength; B3. I propose to assess neural activity associated with item memory and source memory using a novel approach to the ubiquitous problem of task-irrelevant recollection, which has bedeviled previous attempts to study item memory and familiarity in the absence of source memory; B4. I propose to determine the potentially differential effects of mild amnesia, more severe amnesia, and normal forgetting on the two key parameters of recognition (sometimes identified with recollection and familiarity, but which also have a more theory-neutral interpretation). Under topic C (Working memory and the medial temporal lobe), I propose two studies to ask whether the medial temporal lobe is needed for working memory. This issue has been difficult to resolve because of the challenge of knowing when performance on a task actually depends on working memory (or short-term memory) and when it depends on long-term memory. I suggest a novel approach to this issue, using a method suitable for short retention intervals. The proposed studies involve brain structures known to be important for understanding memory and memory impairment as it occurs in a variety of psychiatric and neurological conditions and as a side effect of electroconvulsive therapy. The work can provide new tests, the possibility of better and earlier diagnosis, improve understanding of the conditions that affect memory, and establish a clearer path to the development of interventions for treating and ultimately preventing diseases that affect memory.